
#installation
!pip install pycountry-convert

Requirement already satisfied: pycountry-convert in c:\programdata\anaconda3\lib\site-packages (0.7.2)
Requirement already satisfied: pprintpp>=0.3.0 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (0.4.0)
Requirement already satisfied: repoze.lru>=0.7 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (0.7)
Requirement already satisfied: pytest-mock>=1.6.3 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (3.10.0)
Requirement already satisfied: pytest>=3.4.0 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (6.2.3)
Requirement already satisfied: wheel>=0.30.0 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (0.36.2)
Requirement already satisfied: pytest-cov>=2.5.1 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (4.0.0)
Requirement already satisfied: pycountry>=16.11.27.1 in c:\programdata\anaconda3\lib\site-packages (from pycountry-convert) (22.3.5)
Requirement already satisfied: setuptools in c:\programdata\anaconda3\lib\site-packages (from pycountry>=16.11.27.1->pycountry-convert) (52.0.0.post20210125)
Requirement already satisfied: attrs>=19.2.0 in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (20.3.0)
Requirement already satisfied: iniconfig in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (1.1.1)
Requirement already satisfied: packaging in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (20.9)
Requirement already satisfied: pluggy<1.0.0a1,>=0.12 in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (0.13.1)
Requirement already satisfied: py>=1.8.2 in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (1.10.0)
Requirement already satisfied: toml in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (0.10.2)
Requirement already satisfied: atomicwrites>=1.0 in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (1.4.0)
Requirement already satisfied: colorama in c:\programdata\anaconda3\lib\site-packages (from pytest>=3.4.0->pycountry-convert) (0.4.4)
Requirement already satisfied: coverage[toml]>=5.2.1 in c:\programdata\anaconda3\lib\site-packages (from pytest-cov>=2.5.1->pycountry-convert) (7.0.5)
Requirement already satisfied: tomli in c:\programdata\anaconda3\lib\site-packages (from coverage[toml]>=5.2.1->pytest-cov>=2.5.1->pycountry-convert) (2.0.1)
Requirement already satisfied: pyparsing>=2.0.2 in c:\programdata\anaconda3\lib\site-packages (from packaging->pytest>=3.4.0->pycountry-convert) (2.4.7)

WARNING: Ignoring invalid distribution -portlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -umpy (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -mportlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution - (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -ensorflow-gpu (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -portlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -umpy (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -mportlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution - (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -ensorflow-gpu (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -portlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -umpy (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -mportlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution - (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -ensorflow-gpu (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -portlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -umpy (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -mportlib-metadata (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution - (c:\programdata\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -ensorflow-gpu (c:\programdata\anaconda3\lib\site-packages)
WARNING: You are using pip version 22.0.4; however, version 22.3.1 is available.
You should consider upgrading via the 'C:\ProgramData\Anaconda3\python.exe -m pip install --upgrade pip' command.


import os
os.listdir()

['.ipynb_checkpoints', 'DataSets', 'Insights.ipynb']

pwd

'E:\\DataScience\\MachineLearning\\YouTube Data Insights'


path ='E:\\DataScience\\MachineLearning\\YouTube Data Insights'


dataset =path+"\\DataSets"


os.listdir(dataset)

['List of most-subscribed YouTube channels (1).csv']


# Importing Basics libray
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
sns.set_style('darkgrid')
import plotly
import plotly.express as px
import plotly.io as pio
pio._templates.default ="plotly_dark"
from plotly.subplots import make_subplots
from scipy import signal
from wordcloud import WordCloud ,STOPWORDS

# To Supress Warning
import warnings
warnings.filterwarnings('ignore')

# Map Visulaization
import folium
from folium import Marker
from folium.plugins import MarkerCluster

# To make shell more attractive
from IPython.display import display

# Set Size of Charts 
plt.rcParams['figure.figsize'] =(16,8)
plt.style.use('fivethirtyeight')


df =pd.read_csv(dataset+"\\List of most-subscribed YouTube channels (1).csv")


df.head(10).style.background_gradient(cmap = 'rocket_r')


df.describe().style.background_gradient(cmap = 'rocket_r')


df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 50 entries, 0 to 49
Data columns (total 7 columns):
 #   Column                  Non-Null Count  Dtype  
---  ------                  --------------  -----  
 0   Rank                    50 non-null     int64  
 1   Name                    50 non-null     object 
 2   Brand channel           50 non-null     object 
 3   Subscribers (millions)  50 non-null     float64
 4   Primary language        50 non-null     object 
 5   Category                50 non-null     object 
 6   Country                 50 non-null     object 
dtypes: float64(1), int64(1), object(5)
memory usage: 2.9+ KB


#is there any row having null finally sum up those from each col.
df.isna().sum()

Rank                      0
Name                      0
Brand channel             0
Subscribers (millions)    0
Primary language          0
Category                  0
Country                   0
dtype: int64


df.columns

Index(['Rank', 'Name', 'Brand channel', 'Subscribers (millions)',
       'Primary language', 'Category', 'Country'],
      dtype='object')


# Just making commom lang.
df['Primary language']=df['Primary language'].replace('English[10][11][12]','English').replace('Hindi[13][14]','Hindi').replace('Hindi[16]','Hindi').replace('Hindi[7][8]','Hindi').replace('Hindi[9]','Hindi')


#unique data in each coulmns
df.nunique(axis=0)

Rank                      50
Name                      49
Brand channel              2
Subscribers (millions)    48
Primary language           6
Category                   9
Country                   16
dtype: int64


df['Subscribers (millions)'].nlargest(11)

0     232.0
1     150.0
2     149.0
3     124.0
4     111.0
5     106.0
6     103.0
7      92.4
8      91.9
9      91.0
10     83.6
Name: Subscribers (millions), dtype: float64


# Change the data type to int ..
df['Subscribers (millions)'] = df['Subscribers (millions)'].astype(int)


df_10=df.loc[df['Subscribers (millions)']>83]
df_10.head(10).style.background_gradient(cmap = 'rocket_r')


fig = px.bar(df_10,x=df_10['Name'],y=df_10['Subscribers (millions)'].nlargest(10),
            color=df_10['Name'],title="Top 10 YouTube Channels Which have Highest Numbers of Subscriber",template='plotly_dark')
fig.update_xaxes(title_text='YouTube Channels')
fig.update_yaxes(title_text='Subscribers (millions)')
fig.show()


fig = px.bar(df_10,x=df_10['Brand channel'].value_counts().index,y=df_10['Brand channel'].value_counts(),
            color=df_10['Brand channel'].value_counts().index,title="Top 10 YouTube Channels : count of brand channels",template='plotly_dark')
fig.update_xaxes(title_text='Brand or not')
fig.update_yaxes(title_text='Count')
fig.show()


fig = px.pie(df_10, values=df_10['Category'].value_counts(), names=df_10['Category'].value_counts().index
             , title='Amongs Top 10 Channels : Category %',hole = .65,template='plotly_dark')
fig.show()


fig = px.pie(df_10, values=df_10['Primary language'].value_counts(), names=df_10['Primary language'].value_counts().index
             , title='Amongs Top 10 Channels : Language %',hole = .65,template='plotly_dark')
fig.show()


fig = px.pie(df_10, values=df_10['Country'].value_counts(), names=df_10['Country'].value_counts().index
             , title='Amongs Top 10 Channels : Country %',hole = .65,template='plotly_dark')
fig.show()


fig = px.pie(df, values=df['Brand channel'].value_counts(), names=df['Brand channel'].value_counts().index
             ,title='Brand Presence : %',hole =0.2,template='plotly_dark')
fig.show()


fig = px.pie(df, values=df['Primary language'].value_counts(), names=df['Primary language'].value_counts().index
             ,title='Language : %',hole =0.35,template='plotly_dark')
fig.show()


fig = px.pie(df, values=df['Category'].value_counts(), names=df['Category'].value_counts().index
             ,title='Category : Class',hole =0.35,template='plotly_dark')
fig.show()


fig = px.bar(df,x=df['Country'].value_counts().index,y=df['Country'].value_counts(),
            color=df['Country'].value_counts().index,title="Country With Count of Channel Registered",template='plotly_dark')
fig.update_xaxes(title_text='Country')
fig.update_yaxes(title_text='Counts')
fig.show()


#Preparing data for Map plots
frame = { 'Country': df['Country'].value_counts().index, 'Total':df['Country'].value_counts()}
df_18 = pd.DataFrame(frame)


df_18.reset_index(inplace = True,drop=True)


df_18


df_18.info()
# Country Columns in obj need to convt.

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 16 entries, 0 to 15
Data columns (total 2 columns):
 #   Column   Non-Null Count  Dtype 
---  ------   --------------  ----- 
 0   Country  16 non-null     object
 1   Total    16 non-null     int64 
dtypes: int64(1), object(1)
memory usage: 384.0+ bytes


#Converting object columns to string
df_18['Country'] = df_18['Country'].astype('string')


df_18.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 16 entries, 0 to 15
Data columns (total 2 columns):
 #   Column   Non-Null Count  Dtype 
---  ------   --------------  ----- 
 0   Country  16 non-null     string
 1   Total    16 non-null     int64 
dtypes: int64(1), string(1)
memory usage: 384.0 bytes


#function to convert to alpah3 country codes and continents
from pycountry_convert import  country_name_to_country_alpha3
from pycountry_convert import  country_name_to_country_alpha2
from pycountry_convert import  country_alpha2_to_continent_code


df_18['Country']

0                             India
1                     United States
2                            Brazil
3                       South Korea
4                         Cyprus[a]
5                       El Salvador
6                         Argentina
7                            Russia
8      United States ( Puerto Rico)
9                            Mexico
10                          Romania
11                          Ukraine
12                           Canada
13                   United Kingdom
14                            Chile
15                           Sweden
Name: Country, dtype: string


lst=[]
for cnt in df_18['Country']:
    lst.append(cnt.replace('\xa0', '')) 
lst

['India',
 'United States',
 'Brazil',
 'South Korea',
 'Cyprus[a]',
 'El Salvador',
 'Argentina',
 'Russia',
 'United States(Puerto Rico)',
 'Mexico',
 'Romania',
 'Ukraine',
 'Canada',
 'United Kingdom',
 'Chile',
 'Sweden']


cd=[]
cn=[]
for country in lst:
    #print(country)
    try:
        code =country_name_to_country_alpha3(country)
    except:
        code ='Unknown'
    try:
        code2 =country_name_to_country_alpha2(country)
    except:
        code2 ='Unknown'
    try:
       code1= country_alpha2_to_continent_code(code2)
    except:
        code1 ='Unknown'
        
    cd.append(code)
    cn.append(code1)


cd =pd.Series(cd)
cn=pd.Series(cn)


df_18['CountryCode'] = cd
df_18['ContinentCode'] = cn


df_18


fig = px.scatter_geo(df_18, locations="CountryCode", 
                     color="Total",
                     hover_name="Country", 
                     size="Total",
                     title = 'Distribution of Registered YouTube Channels using Bubble Maps',
                     projection="natural earth",
                     color_continuous_scale =px.colors.sequential.Rainbow,
                     template='plotly_dark'
                    )
fig.show()


fig = px.choropleth(df_18,
                    locations = "CountryCode",
                    color = "Total",
                    hover_name = "Country",
                    color_continuous_scale =px.colors.sequential.Rainbow,
                    projection = 'orthographic',
                    title = 'Distribution of Registerd Youtube Channels in World Map',
                    height = 600,
                    width = 1000,
                    template='plotly_dark'
                   )
fig.update_geos(lataxis_showgrid = True,
                lonaxis_showgrid = True,
                showcountries = True,
               )

fig.update_geos(lataxis = {'gridcolor':'#222222'},
                lonaxis = {'gridcolor':'#222222'},
               )
fig.show()


df_18['ContinentCode'].value_counts()

EU         5
NA         4
SA         3
AS         2
Unknown    2
Name: ContinentCode, dtype: int64


fig = px.bar(df_18,x=df_18['ContinentCode'].value_counts().index,y=df_18['ContinentCode'].value_counts(),
            color=df_18['ContinentCode'].value_counts().index,title="continent Wise YouTube Channels",template='plotly_dark')
fig.update_xaxes(title_text='continent')
fig.update_yaxes(title_text='Count')
fig.show()


## Co-relations
import phik 
from phik.report import plot_correlation_matrix
from phik import report


phik_overview= df.phik_matrix()

interval columns not set, guessing: ['Rank', 'Subscribers (millions)']


phik_overview.style.background_gradient(cmap = 'rocket_r')


# Author ~ Saurabh Kumar..........xxxxxxxxxxxxxxxxxxxxxxxxx

	Rank	Name	Brand channel	Subscribers (millions)	Primary language	Category	Country
0	1	T-Series	Yes	232.000000	Hindi[7][8]	Music	India
1	2	Cocomelon	Yes	150.000000	English	Education	United States
2	3	Sony Entertainment Television India	Yes	149.000000	Hindi[9]	Entertainment	India
3	4	MrBeast	No	124.000000	English	Entertainment	United States
4	5	PewDiePie	No	111.000000	English	Entertainment	Sweden
5	6	Kids Diana Show	Yes	106.000000	English[10][11][12]	Film	Ukraine
6	7	Like Nastya	No	103.000000	English	Entertainment	United States
7	8	WWE	Yes	92.400000	English	Professional wrestling	United States
8	9	Vlad and Niki	No	91.900000	English	Entertainment	Russia
9	10	Zee Music Company	Yes	91.000000	Hindi[13][14]	Music	India

	Rank	Subscribers (millions)
count	50.000000	50.000000
mean	25.500000	70.442000
std	14.577380	34.660653
min	1.000000	44.000000
25%	13.250000	50.050000
50%	25.500000	57.500000
75%	37.750000	77.475000
max	50.000000	232.000000

	Rank	Name	Brand channel	Subscribers (millions)	Primary language	Category	Country
Rank	1.000000	0.934369	0.102468	0.764759	0.667582	0.080475	0.574447
Name	0.934369	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
Brand channel	0.102468	1.000000	1.000000	0.122210	0.703057	0.290028	0.851007
Subscribers (millions)	0.764759	1.000000	0.122210	1.000000	0.000000	0.000000	0.000000
Primary language	0.667582	1.000000	0.703057	0.000000	1.000000	0.000000	0.934360
Category	0.080475	1.000000	0.290028	0.000000	0.000000	1.000000	0.302373
Country	0.574447	1.000000	0.851007	0.000000	0.934360	0.302373	1.000000

Importing Libray¶

Top 10 YouTube Channels¶

STORY ~ INSIGHTS¶

TOP 10 channels by highest number of subscriber¶

Out of top 10 channels which are belongs to brand ?¶

Amongs Top 10 Channels : Category %¶

Amongs Top 10 Channels : Language %¶

Amongs Top 10 Channels : Country %'¶

Obseravation ~¶

Now Let us See Big-Picture ~👀¶

Brand Presence %¶

Prevalence of the language %¶

Popular Categories Class¶

Country With Count of Channel Registered¶

Distribution of Registered YouTube Channels using Bubble Maps¶

Obseravations ~¶

! Thanks EveryOne I hope you enjoyed this notebook , Just Let me know if you have any feedback ,will HAPPY to See your FeedBacks ..~👀¶

	Rank	Name	Brand channel	Subscribers (millions)	Primary language	Category	Country
0	1	T-Series	Yes	232	Hindi	Music	India
1	2	Cocomelon	Yes	150	English	Education	United States
2	3	Sony Entertainment Television India	Yes	149	Hindi	Entertainment	India
3	4	MrBeast	No	124	English	Entertainment	United States
4	5	PewDiePie	No	111	English	Entertainment	Sweden
5	6	Kids Diana Show	Yes	106	English	Film	Ukraine
6	7	Like Nastya	No	103	English	Entertainment	United States
7	8	WWE	Yes	92	English	Professional wrestling	United States
8	9	Vlad and Niki	No	91	English	Entertainment	Russia
9	10	Zee Music Company	Yes	91	Hindi	Music	India

	Country	Total
0	India	17
1	United States	12
2	Brazil	4
3	South Korea	4
4	Cyprus[a]	2
5	El Salvador	1
6	Argentina	1
7	Russia	1
8	United States ( Puerto Rico)	1
9	Mexico	1
10	Romania	1
11	Ukraine	1
12	Canada	1
13	United Kingdom	1
14	Chile	1
15	Sweden	1

Importing Libray¶

Top 10 YouTube Channels¶

STORY ~ INSIGHTS¶

TOP 10 channels by highest number of subscriber¶

Out of top 10 channels which are belongs to brand ?¶

Amongs Top 10 Channels : Category %¶

Amongs Top 10 Channels : Language %¶

Amongs Top 10 Channels : Country %'¶

Obseravation ~¶

Now Let us See Big-Picture ~👀¶

Brand Presence %¶

Prevalence of the language %¶

Popular Categories Class¶

Country With Count of Channel Registered¶

Distribution of Registered YouTube Channels using Bubble Maps¶

Obseravations ~¶

! Thanks EveryOne I hope you enjoyed this notebook , Just Let me know if you have any feedback ,will HAPPY to See your FeedBacks ..~👀¶

If you enjoyed just upvote the kernal and share with those who likes these stuff..¶